don't break while generating tokens for names with non-ascii characters

[davisagli]

(which could come from user input in the creator field,
see plone/plone.app.widgets#120)

[davisagli]

Tests: http://jenkins.plone.org/job/pull-request-5.0/600/

[davisagli]

Trying again: http://jenkins.plone.org/job/pull-request-5.0/601/

[davisagli]

Fixed test, trying again: http://jenkins.plone.org/job/pull-request-5.0/603/

[davisagli]

Passing

[jensens]

fetching fullname can be expensive, did you try this with many users?

[davisagli]

I didn't change that part, I just made it encode the value to create the token when it needs to, and reduced code duplication.

[thet]

looks good, only i'm wondering why you are using the unicode_escape encoding for the token and not quoted-printable encoding, like for plone.app.vocabularies.catalog KeywordsVocabulary?

[jensens]

I'd just merge, i think which encoding is used here does not make a big difference. I'd have used quoted-printable too, since its an encoding which can be used outside python world too easily.
But we may want to change this in another PR if this is a thing.

[thet]

+1

[davisagli]

I used unicode_escape out of habit...I wasn't really familiar with quoted-printable, but it looks like a good choice too.